Risk factors for endometrial cancer, including weight and a high-fat diet, operate through estrogen-related mechanisms. Estrogen promotes growth of transformed cells and there is evidence that some metabolites may have carcinogenic properties. Some of the genes that are important in estrogen metabolism have been found to be polymorphic (CYP17, CYP1 Al, catechol 0-methyltransferase, CYP1 Bi, and CYP19), raising the question of whether genotypes are related to risk. The primary aim of this study is to compare endometrial cancer cases to controls to determine the associations of weight, diet, and genetic susceptibility, independently and jointly, with endometrial cancer. Most endometrial tumors are of endometrioid histology. The rarer serous and clear cell tumors are more aggressive and lethal. Aside from their higher prevalence in older women and in blacks, little is known about risk factors for these tumors, although they appear not to be related to estrogen. The secondary aim is to compare cases with these tumor types to controls to assess risk factors for these poorly-understood tumors. We will conduct a population-based case-control study in six counties in New Jersey. Cases will be women with newly-diagnosed endometrial cancer. Controls will be matched by 5-year age groups and selected by random digit dialing (for those under 65) or from HCFA files (for those aged 65 and over). There will be 600 cases with endometrioid tumors, 200 with serous or clear cell tumors, and 600 controls. Endometrial cancer, already a serious cause of morbidity in the US, is likely to become more common in future years as the population ages and obesity increases. Risk factors such as obesity and poor diets are potentially modifiable. This study will provide data on genetic susceptibility in endometrial cancer in conjunction with established risk factors, as well as the first epidemiologic data on the more lethal serous and clear cell tumors.